These studies are designed to obtain information about drug delivery in experimental brain tumors. Our previous work has shown low rates of blood-to-tissue transport of water soluble compounds in experimental brain tumors. The specific aims are: 1) to study the effects of corticosteroids and x-irradiation on blood flow and blood-to-tissue transport; 2) to study methods that may increase the rate of blood-to-tissue transport in brain tumors (hyperosmotic disruption, hypertension, hypercapnia, hyperthermia and dimethyl sulfoxide); 3) to develop an in vivo method to measure the rate of blood-to-tissue transport using computed tomographic (CT) scanning; 4) to measure the rates of blood-to-tissue transport of several commonly used cancer chemotherapeutic drugs, and; 5) to corrrelate the rate of blood-to-tissue transport with ultrastructural features of capillaries and measure the surface area of capillaries available for transcapillary exchange. Blood flow will be measured with Kety-Schmidt equations. The rate of blood-to-tissue transport will be measured with alpha-aminoisobutyric acid (which has unidirectional blood-to-tissue transport) or with the slope-intercept method of data analysis for compounds with bi-directional capillary transport. Tissue concentrations of radiolabeled drugs will be measured by quantitative autoradiography. The surface area of tumor capillaries will be measured after staining the capillaries with Factor VIII/von Willebrand antiserum and peroxidase-antiperoxidase. Since most chemotherapeutic drugs are water soluble, and since water soluble compounds have low rates of transcapillary transport in brain tumors, we hope to find methods with which drug delivery can be increased to brain tumors. In order for these methods to be applied to humans, a method to measure the rate of transcapillary transport in vivo will be needed, for which we plan to use the CT scanner and the slope-intercept method of graphical analysis. We also plan to study the transport properties of brain tumor capillaries to see if the rate of transcapillary transport of chemotherapeutic drugs can be predicted, based on either the water solubility or molecular size of the drug. If these objectives can be accomplished, they will have direct applicability in the treatment of human brain tumor patients.